Rethinking Admission Location for Low Acuity Infants of 35 Weeks’ Gestation

Late preterm infants born at 35 weeks’ completed gestational age (GA) account for >75 000 infants born in the United States annually.^1,2 Although prematurity places them at higher risk of complications, such as immature feeding, impaired thermoregulation, or hypoglycemia, many infants born at 35 weeks GA are clinically stable without the need for intervention.^3,4 Accordingly, whereas a subset of these infants may require a higher level of care in a NICU, others may be unnecessarily missing out on the benefits of colocation within a mother–baby unit (MBU). Colocation with a parent reduces the risk of infection and improves breastfeeding, bonding, and parental mental health.^5–7 Despite these known benefits, little evidence is available to guide infant placement decisions, including the identification of infants most likely to benefit from admission to a NICU.

In this issue of Pediatrics, Wickremasinghe et al⁸ evaluated outcomes of admission location for infants born at 35 weeks’ GA within an integrated health system. This study used electronic health record data from a retrospective cohort of 5929 low acuity infants born at 35 0/7 to 35 6/7 weeks’ GA from 2011 to 2021 across 13 Kaiser Permanente Northern California hospitals. Low-acuity infants were those for whom NICU admission was likely due to hospital policy rather than medical necessity. Chart data were analyzed to compare infant length of stay, breastfeeding rates, and readmission rates according to admission location.

The primary outcome, length of stay during the birth hospitalization, was longer for infants admitted to a NICU (58 hours longer in adjusted analyses, 95% confidence interval [CI] 51–64). Lower exclusive breast milk feeding during the birth hospitalization (3% vs 10%, adjusted odds ratio [aOR] 0.44, 95% CI 0.28–0.69) and at 6 months by caregiver report (15% vs 25%, aOR 0.73, 95% CI 0.55–0.97) was also observed in the NICU group. Exclusive breast milk feeding at 2 months was not statistically significantly different according to infant placement. Direct breastfeeding versus the provision of pumped milk was not specified in this cohort.

An expected consequence of shorter birth hospitalizations is a higher readmission rate due to indications such as hyperbilirubinemia or immature feeding, which might not present within the first days of life.⁹ Indeed, the infants directly admitted to a NICU had fewer readmissions (3% vs 6%, aOR 0.43, 95% CI 0.27–0.69). This association was no longer statistically significant when comparing only readmissions occurring later than the mean NICU stay of 96 hours, suggesting that some but not all readmissions may have been averted with a longer birth hospitalization. The higher readmission rate in the MBU group was true across nearly all categories of readmission diagnoses, including common late preterm conditions, but also other potentially acutely serious diagnoses, such as respiratory, gastrointestinal, and cardiac disorders. An important future direction will be to disentangle the relative contributions of length of stay and level of care to the risk of readmission, receipt of timely treatment, and subsequent outcomes.

In our clinical experience, readmissions can be frightening and disheartening for new families. They can result in invasive evaluation and/or family separation, and they can also be costly. At the same time, prolonging birth hospitalizations in a large number of infants to prevent readmission in a small few has its own implications for families and for health systems. In optimizing the tradeoff between birth hospitalization duration and readmission risk, clinicians should ensure that infant safety is not compromised, acknowledging that not all readmissions are considered “failures” of the system.

This study addresses a critical gap in our understanding of the implications of admission location for infants born at 35 weeks’ GA. Birth hospitalization admission decisions and policies have potentially far-reaching ramifications on clinical care and outcomes, patient experience, and the cost and allocation of health care resources. Providing evidence for the scope of effect, our recent study conducted through the Better Outcomes through Research for Newborns network revealed wide variation in admission practices across a representative sample of US hospitals, with 10% of hospitals requiring NICU admission for infants born at 35 weeks’ GA.¹⁰ Based on the findings of Wickremasinghe et al, admitting 100 infants to an MBU versus a NICU would result in 242 fewer birth hospitalization days and 5 additional infants who are exclusively breastfed at 6 months, although with 3 to 4 infants being readmitted.

Newborn providers caring for low-acuity late preterm infants would appreciate the identification of infant characteristics present early in the birth hospitalization that predict favorable outcomes, which may help guide admission and discharge decisions and maximize the number of dyads who can safely receive all the benefits of colocation. Future postnatal care units and staffing programs can furthermore be designed to meet the unique needs of late preterm infants and their families.

Glossary

aOR

adjusted odds ratio

CI

confidence interval

GA

gestational age

MBU

mother–baby unit